Twenty Second Meeting of the IMF Committee on Balance of Payments Statistics Shanghai, China P.R., November 2 4, 2009

Transcription

1 BOPCOM-09/08 Twenty Second Meeting of the IMF Committee on Balance of Payments Statistics Shanghai, China P.R., November 2 4, 2009 Strategy for Implementing the Sixth Edition of the Balance of Payments and International Investment Position Manual Prepared by the Statistics Department International Monetary Fund

2 2

3 3 TABLE OF CONTENTS I. Introduction... 4 II. Migration from BPM5 to BPM6 in STA publications... 5 Target date for the conversion to BPM Remapping BPM6 data to BPM5 prior to the conversion date... 6 Remapping BPM5 data to BPM6 after the conversion date... 6 Historical and overlapping data after the conversion date... 6 Major effects of the remapping... 8 Database and external codes... 9 III. Balance of Payments and International Investment Position Compilation Guide... 9 A. Structure... 9 B. Work plan IV. Training Program For BPM

4 4 I. INTRODUCTION 1. With the adoption of the sixth edition of the Balance of Payments and International Investment Position Manual (BPM6), the IMF is now focusing on assisting member countries in implementing the updated standards. The broad outline of a strategy was presented to the Committee in 2007 and further details were outlined in This paper reviews the steps that have been taken since the Committee s last meeting, and discusses actions that the IMF and the Committee may undertake in the coming year in continuation of implementation goals. 2. The Committee previously endorsed the proposal for an implementation strategy that includes updating the BPM6 Compilation Guide (BPM6 CG) and providing assistance to balance of payments compilers in the form of training and seminars. The Committee s views are consistent with those expressed at the regional outreach seminars on the implementation of the BPM6. A number of participants at those seminars remarked that the timely completion of the BPM6 CG could facilitate the implementation of the updated standards. 3. The Committee has also recognized that a key element of the implementation strategy is the timely changeover of the presentation of data in the IMF s Balance of Payments Yearbook (BOPSY) and the International Financial Statistics (IFS) to the BPM6 format as of a fixed date, which the IMF has set as 2012 (for 2011 and possibly earlier years data). 4. In regard to other elements of the implementation strategy, the IMF has completed work on a conversion matrix, showing the correspondence between the BPM5 and the BPM6 accounts. Also, the IMF has conducted a questionnaire of all member countries to inquire about their implementation plans and any challenges encountered. A purpose of the questionnaire was to identify areas where compilers may appreciate IMF assistance in converting to the new standards. 1 Work on the BPM6 CG has begun. 5. The remainder of this paper is organized as follows: Section II outlines the IMF s plans to convert presentations in BOPSY and IFS to the BPM6 format; Section III reviews plans to update the BPM6 Compilation Guide (BPM6 CG); and Section IV reviews the IMF s training programs that have been targeted toward adoption of BPM6. An outline for the BPM6 CG is presented in Annex I. 1 The results from the questionnaire are presented in detail in BOPCOM-09/04.

5 5 II. CONVERSION FROM BPM5 TO BPM6 PRESENTATIONAL BASIS IN STA PUBLICATIONS 6. The IMF Statistics Department (STA) collects balance of payments and International Investment Position (IIP) data from its member countries. The data are redisseminated in several forms, including hardcopy publications, CD-ROM, and on-line databases. STA will adapt its current data production and dissemination tools and practices, to present data consistent with the standard components in BPM6. 7. STA is in the process of more fully articulating its plans for adopting the BPM6 presentation in these publications. This work includes adopting a strategy to address conversion issues arising during what may be a prolonged interim period, during which some countries will have partially or fully implemented BPM6, and others will still be fully on a BPM5 basis. Staff from STA s Balance of Payments Division and the Statistical Information Division have been working together to address this and other implementation issues in all STA publications. Target date for the conversion of presentations to BPM6 8. Following consultations with compilers worldwide and with the Committee, the IMF has set 2012 (with reference to data for 2011) as the target date for converting the balance of payments data to the BPM6 standard presentation format. Data will therefore be presented as follows: (i) according to the BPM5 presentational basis before the conversion date in 2012; (ii) according to both the BPM5 and BPM6 presentational bases for an overlapping period after the 2012 conversion date (perhaps only for economies that continue to report data to STA on a BPM5 basis); and (iii) according to the BPM6 for all years starting with By publishing data on both a BPM5 and a BPM6 basis for a period of time, the data in the IMF s statistical publications will match those disseminated on national websites of its member countries; in addition, this can help data users assess the impact of adopting the updated standards. 9. It should be noted that the 27 members of the European Union are required to convert their presentations to a BPM6 basis in 2014 (with reference to data for 2013 and perhaps for some earlier years). These efforts in Europe should dovetail well, because the IMF s work in developing software, conversion matrixes, technical discussions with compilers, and other outreach efforts will enable countries in the European Union to develop pilot estimates on a BPM6 basis before their mandated conversion to a BPM6 basis in In light of the heavy demands on these countries to convert all of their data sets (external, national accounts, monetary and financial, and government finance) to the new standards all at the same time, the work that the IMF is performing in designing data conversion tools and techniques should alleviate a portion of the burden.

6 6 Converting BPM6 data to a BPM5 basis prior to the 2012 conversion date 10. STA intends to convert data that are reported on a BPM6 basis before 2012 back to a BPM5 basis, using in-house tools and supplementary information provided by the countries. Nonetheless, the IMF intends to provide a BPM6 reporting tool, comprising all the BPM6 standard components and supplementary items, for use by countries that have converted to BPM6 early, to facilitate their reporting of information to the IMF. The tool may include flags and country notes, to identify breaks in series (from BPM5 to BPM6). The reporting tool may also include questions that countries would answer that would assist STA in converting BPM6 data back to a BPM5 basis. Converting BPM5 data to BPM6 basis after the 2012 conversion date 11. STA intends to convert data that are reported on a BPM5 basis in 2012 and later years to the BPM6 basis. Thus, STA will need to provide a BPM5 reporting tool (as well as the previously discussed BPM6 reporting tool). Converting data accurately from a BPM5 basis to a BPM6 basis may be more difficult than converting from a BPM6 basis to a BPM5 basis, because compilers may not be able to confidently judge the impact of the conversion to BPM6 in advance of actually adopting the new standards. 12. The conversion may entail the following: (i) automated routines and standard rules may be developed to convert BPM5 items that can be linked directly to BPM6 items; (ii) for other items, conversion ratios may be developed in consultation with the reporting economy, and perhaps can be incorporated in the reporting tool; (iii) information may be collected from representative groups of reporting countries that have converted to BPM6 that would then be shared with other countries to facilitate their data conversion efforts. We will inform data users of data that have been converted, and may flag the converted data, i.e., identify which data were directly compiled and reported by countries, and which data were developed with country involvement through automated routines, customized conversion rules, and other techniques. Historical data after the conversion date 13. Many countries may find it challenging to convert BPM5 data for a very long historical period to a BPM6 basis, due to large changes that may occur in some data series. Further consultations with the Committee and compilers are needed to obtain input regarding the length of a time period for which historical data should be converted to a BPM6 basis. The choice may be between accepting a degree of imprecision in the BPM6 basis estimates for the historical time series, versus having a short historical time series. In past conversions of BOP/IIP data from one version of the Balance of Payments Manual to the next, the full historical time series (decades of data) was converted to the updated basis. There may be a substantial demand for a lengthy historical time series by data users within and outside the

7 7 IMF, and indications are that imprecise estimates for the historical period would be tolerated unless these estimates would be highly misleading at fairly aggregate levels. Overlapping years, when data are presented on both a BPM5 and BPM6 basis 14. STA is exploring providing data on both a BPM5 and a BPM6 basis for some number of overlapping years. Consideration is being given to maintaining the BPM5 time series through 2014 (when members of the European Union are required to convert to a BPM6 basis), and to commence the BPM6 basis time series in That is, the IMF is exploring the feasibility of publishing data on both a BPM5 and a BPM6 basis for one or more years, perhaps only for those countries that continue to report BPM5 basis data after Under this approach, the BPM5 time series may not end in 2011 (for 2010 data) but would instead be continued through 2012 (for 2011 data) or a later year (not later than 2014). 15. BPM6 data would also be published, starting in 2012 (for 2011 data). The BPM6 basis data series may also cover the entire historical time series that is also covered by BPM5 basis data. The BPM6 basis data would extend into the future and the BPM5 basis data would terminate as of a specified date (2011, which would mean no years of overlap, or a later year, in ). The BPM5 basis data may not be continually maintained by the IMF; that is, these data would not necessarily be routinely revised, so that the IMF is not routinely publishing two sets of estimates. If data are released on a BPM5 basis during an overlapping period, they would probably be available only on-line and via magnetic media, and not in hard copy publications as well. Producing and maintaining data on both a BPM5 and BPM6 basis for an overlapping period of time (whether for 2012 through 2014 or for a briefer period) carries significant resource implications for STA, and availability of resources for this work is not assured. 16. Under the above approach, beginning in 2012, STA will convert 2011 data that are reported according to a BPM5 basis to a BPM6 basis, and during some overlap period, STA may convert data that are reported according to a BPM6 basis to a BPM5 basis, so that data on both bases would be available for a period of time, resources permitting. In summary, STA is considering the following procedure for the conversion from the BPM5 to the BPM6 data presentation in its publications: Until the conversion of its statistical publications in 2012, STA will publish only BPM5 basis data for all economies. More specifically, data that are reported to STA on a BPM6 basis would not be redisseminated by STA until 2012 and later years; instead, through 2011, data that STA receives on a BPM6 basis would be converted to a BPM5 basis before publication.

8 8 After the conversion of its statistical publications in 2012, STA will publish data in BPM6 format for all countries. A decision will need to be taken regarding whether to convert BPM5 basis data for an historical period to a BPM6 basis. Data that are reported to STA on a BPM5 basis in 2012 and later years (pertaining to data for 2011 and later years) will be converted to a BPM6 basis. After the conversion of its statistical publications in 2012, resources permitting, STA may publish data on both a BPM5 and BPM6 basis for all countries for an overlapping period. The length of this overlapping period needs to be considered. After the conversion of its statistical publications in 2012, STA may publish the BPM5 data series in electronic format only (on CD-ROM and the on-line database). In this circumstance the hardcopy publications (BOPSY and IFS) would contain an historical series for BPM6 basis data (or would contain a break in series). It would likely cause data users some problems if, for example, data referring to earlier years are available only on a BPM5 basis and data referring to later years are available only on a BPM6 basis, suggesting that production of an historical time series for BPM6 basis data should be carefully considered. Major effects of converting to BPM6 17. The impact of converting to BPM6 may be significant, especially in the following cases where major account balances for some countries could be significantly impacted. Goods account: reclassifications for goods for processing, repair of goods and durable goods bought by travelers. Services account: mirror effects of reclassifications in goods account; FISIM; and insurance services (reclassifications between services and primary income, secondary income, and financial account). Primary income, secondary income, financial account: mirror effects of reclassifications in FISIM and financial services in the services account. Capital account: changes due to exclusion of migrants transfers from transactions, and mirror effects in financial account and in trade in goods. Direct investment: gross presentation of assets and liabilities, instead of netting in accordance with the directional principle. 18. STA will assess, or will assist in the assessment of, the implications for other datasets, e.g., world tables, IMF quota formula, and WEO tables.

9 9 Database and external codes 19. STA will review the internal database and external reporting tool codes. In regard to the external codes, STA s work will be performed in cooperation with other international agencies. III. BALANCE OF PAYMENTS AND INTERNATIONAL INVESTMENT POSITION COMPILATION GUIDE A. Structure 20. At the Committee s 21 st meeting, the IMF presented key elements of an approach for updating the BPM6 CG. The IMF plans to use a modular approach by producing and disseminating the BPM6 CG in separate chapters with priority given to methodological changes that are especially difficult to implement. Committee members supported the proposal to treat the BPM6 CG as a living document, and to publish updated chapters and new boxes in electronic form upon their completion. The Committee also agreed with a request to give attention to the backcasting of data in the BPM6 CG. Further, there was a proposal to coordinate the update of compilation issues that affect both BPM6 and the 2008 SNA. The Committee noted that it would be useful to have contributions from Committee members and others to address issues common to BPM6 and the 2008 SNA. 21. The IMF has prepared a broad outline for the BPM6 CG, which is based on the structure of the BPM5 Compilation Guide. For purposes of drafting, the BPM6 CG can be considered in three sections. 22. The first section will examine the broad groups of data sources. Each chapter will review the relative advantages and disadvantages of the data source in terms of the conceptual consistency with the BPM6, the resource requirements (human, financial, and other resources), and the timeliness of the data. This section is predicated on the assumption that one data source may be used to develop estimates of various items; therefore, except in a few instances, the data sources will not be discussed in detail with specific reference to BOP/IIP items. 23. The second section will review the compilation of the various items of the balance of payments and IIP statistics in terms of the data requirements and the statistical techniques that could be used to estimate these items. These techniques include simple estimation, sample expansion, the use of data models, and extrapolation/interpolation. The section will also examine each method in terms of data quality considerations (the accuracy, reliability, and the timeliness of account estimates). It will note the data sources (listed in the first section) that underpin these statistical techniques.

10 The third section will review a range of general issues, such as the presentation of the data, and issues relating to the recording of transactions relating to exceptional financing. It will also cover additional issues relating to data quality such as recommended timeliness, frequency, and dissemination practices. These data quality issues will be discussed within the context of the IMF s Data Quality Assessment Framework 2. In addition, this section will include chapters on special topics and crosscutting issues that users have generally expressed difficulty in understanding. Thus, the BPM6 CG will include a chapter on insurance flows and positions that will examine the range of crosscutting issues relating to insurance. 25. The BPM6 CG will examine in greater detail, issues relating to the compilation of IIP statistics. It will therefore incorporate the substantive areas of material currently contained in International Investment Position: A Guide to Data Sources. Some other sections of the BPM6 CG are not expected to be as comprehensive as the previous compilation guide as the BPM6 now addresses many issues that were addressed in the previous compilation guide. Throughout, the BPM6 CG will place special emphasis on addressing compilation issues relating to the major changes identified in the BPM6 Appendix In addition, the BPM6 CG will advance the work on direct investment compilation issues that have been noted. Indeed, the area of direct investment has been repeatedly identified as posing special challenges in the implementation questionnaire that the IMF recently conducted (see BOPCOM 09/04). The IMF is using results from the questionnaire to guide its work on the BPM6 CG. Some areas where compilers have indicated they would appreciate receiving compilation help from the IMF include the treatment of fellow enterprises, identifying Ultimate Controlling Parents, and statistical unit issues (see BOPCOM 09/12). 27. The BPM6 CG will be closely aligned with other existing compilation guides such as the Coordinated Portfolio Investment Statistics Guide, the Coordinated Direct Investment Survey Guide, the Remittances Compilation Guide, and the External Debt Statistics Compilation Guide. The BPM6 CG will refer to these other guides where necessary, and will avoid replicating large areas of the material from these guides. 28. STA will encourage the input of compiling agencies in drafting sections of the BPM6 CG. The BPM6 CG will include examples of country practices in both data collection and estimation. It will include examples from a range of countries in an attempt to show how countries address or overcome problems in data collection or estimation. It will also comment on the implications for the accuracy and reliability of the estimates. The examples 2 There currently is no data quality assessment framework for the IIP; however, the principles of the framework could be usefully applied to the dataset.

11 11 will therefore be drawn from both developed and developing countries with varying degrees of openness and degrees of institutional development. B. Work plan 29. STA will establish a task force to provide input that it can adapt for use in the BPM6 CG. The task force will include IMF staff as well as representatives from individual countries and international organizations with expertise in the topic and interest in contributing. The work is expected to be conducted electronically through the exchange of documents rather than through face-to-face meetings. The IMF has previously had interactions with the World Tourism Organization (UNWTO) on contributing to the sections of the BPM6 CG relating to travel and tourism statistics. The impetus for this work is based on longstanding collaboration between the IMF and UNWTO on issues relating to statistics on travel and tourism. The contributions of country representatives will not be restricted to any particular area of the BPM6 CG; however, it is expected that these contributions may focus on areas relating to data collection and estimation. 30. In drafting the BPM6 CG, special emphasis will be placed on the issues and areas identified by compilers during the regional outreach seminars as urgent or difficult to implement. The results of the worldwide survey on the implementation of the BPM6 also provide some additional insights that compilers wish to see reflected in the BPM6 CG. 31. STA with the support of its task force should produce a first draft of most of the chapters of the BPM6 CG for comments in time for the next meeting of the Committee (which may be held at end-october 2010). Draft chapters will be placed on a special page created on the IMF website for comment on a continuous basis, as they become available. That is, the website will be used to publish the living document, which will be updated and augmented frequently. The BPM6 CG is expected to be finalized and published (at least) in electronic form in Regional seminars IV. TRAINING PROGRAM FOR BPM6 32. In 2008, STA launched a series of regional outreach seminars to present the BPM6 to member countries. STA conducted nine one-week regional outreach seminars to introduce the main changes from BPM5 to senior staff in charge of balance of payments and IIP statistics from 173 economies. Efforts to encourage countries to implement the BPM6 were advanced with the revision of all regular balance of payments training materials to incorporate the concepts, guidelines, and recommendations of the BPM6. All course materials for the regional courses are based on BPM6. To date, STA has covered all regions some on multiple occasions with courses based on the BPM6 (see Table). As

12 12 noted, the feedback received from discussions during the seminars and courses conducted during 2008 and 2009 is being used in conjunction with the information from the worldwide survey on BPM6 implementation to identify the areas that should be addressed in the update of the BPM6 CG. Courses on External Sector Statistics in * (Number of courses, fiscal years May-April) Number of Courses Region 2008/ /2010 AFR (Tunisia, The Gambia) 3 2 APD (China, India, Singapore) 2 5 EUR (Austria, Slovenia) 0 3 MCD (United Arab Emirates) 1 0 WHD (Brazil) 2 1 Headquarters (Washington, D.C.) 1 2 TOTAL 9 13 *Courses on BPM6, including topical courses on remittances and services. Excluding courses on CDIS, CPIS; and one-week outreach seminars. 33. STA has also undertaken a series of workshops to promote the Coordinated Direct Investment Survey. These workshops have been in the form of large regional groups as well as bilateral discussions with individual countries. This exercise is expected to increase the

13 13 number of countries participating in the survey and also enhance the pace of BPM6 implementation. 34. In addition to the broad courses in balance of payments and IIP statistics, STA will be conducting a series of shorter (3-5 days) courses in topical areas. These include courses in remittances statistics, and statistics on international trade in services, at the Center of Excellence in Finance in Slovenia, or at the IMF s Singapore Training Institute. IMF Headquarters Courses 35. All lecture presentations, workshop exercises, and case studies for the course at the IMF Headquarters (Headquarters Course) have been fully converted to BPM6. Feedback from participants has been taken into account for next year s Headquarters Course. Participants recommended more extensive coverage of complex issues, such as insurance and pension fund transactions and financial derivatives. Training of IMF staff 36. The IMF has also instituted a program of training in BPM6 for IMF macroeconomists. The training, which will be conducted under the IMF Institute s Economist Training Program, will be tailored according the needs of the IMF s area departments 3. The training of Fund economists will provide an additional channel through which the use of the BPM6 in member countries can be promoted. Training of IMF experts 37. The IMF maintains a cadre of experts who provide technical assistance on behalf of the IMF. The experts have received copies of the BPM6, the conversion matrix, and additional background material. Questions for the Committee: 1. The members of the Committee are invited to comment on the draft outline for the BPM6 CG (see Annex I). 2. Do Committee members agree with the work plan for completing the BPM6 CG? (See paragraphs ) 3 Five departments within the Fund that are responsible for economic surveillance. These departments are based on regional groupings of countries (African Department, Asia and Pacific Department, European Department, Middle East and Central Asia Department, and Western Hemisphere Department).

14 14 3. Would members of the Committee be willing to contribute to the BPM6 CG? Could members indicate the areas of the BPM6 CG on which they would like to collaborate? 4. Do members of the Committee have views at this time on the length of the historical time period over which BPM6 basis data should be produced? (See paragraph 16. This question may be revisited at subsequent Committee meetings.) 5. Do members of the Committee have views on whether STA should publish data on both a BPM5 and BPM6 basis for an overlapping period, after the conversion to BPM6 in 2012? If yes, for what period should there be an overlap? Finally, should estimates be disseminated for all countries, or only for countries that continued to disseminate their data on a BPM5 basis? 6. Can members of the Committee help with setting or reviewing data conversion rules (from BPM5 to BPM6, or from BPM6 to BPM5)?

15 ANNEX I OUTLINE BALANCE OF PAYMENTS AND INTERNATIONAL INVESTMENT POSITION MANUAL COMPILATION GUIDE The Balance of Payments and International Investment Position Compilation Guide (BPM6 CG) will be a companion document to the sixth edition of the Balance of Payments and International Investment Position Manual (BPM6). It will therefore update the current Balance of Payments Compilation Guide that accompanies the BPM5. Compared to the previous version of the compilation guide, the BPM6 CG will examine in greater detail, issues relating to the compilation of international investment position (IIP) statistics. It will therefore incorporate a large amount of the material currently contained in International Investment Position: A Guide to Data Sources. Some other sections of the BPM6 CG are not expected to be as comprehensive as its predecessor as the BPM6 now addresses many issues that were addressed in the previous CG. The BPM6 CG will also place special emphasis on addressing compilation issues relating to the major changes identified in the BPM6 Appendix 8. The BPM6 CG will be in three sections. 4 Section A will examine the broad groups of data sources. Each chapter will review the relative advantages and disadvantages of the data source in terms of the conceptual consistency with the BPM6, the resource requirements (human, financial and other resources), and the timeliness of the data. This section is predicated on the assumption that one data source may be used to develop estimates of various items; therefore, except in a few instances, the data sources will not focus on specific balance of payments and IIP items. Section B will review the compilation of various balance of payments and IIP accounts in terms of the data requirements and the statistical techniques used to estimate the items. These techniques include simple estimation, sample expansion, the use of data models, and extrapolation/interpolation. The section will review the range of data sources and the range of techniques that could be used for each item. The section will also examine methods in terms of data quality considerations (accuracy, reliability, and timeliness of estimates). Section C will review a range of general issues such as the presentation of the data and the development of the analytic summary. The BPM6 CG will include examples of country practices in both data collection and estimation. It will include examples from a range of countries in an attempt to show how countries address or overcome some of the many problems of data collection and their implications for the accuracy and reliability of the estimates. The examples will therefore be 4 The delineation by section is for purposes of drafting and may not be reflected in the BPM6 CG.

16 16 drawn from both developed and developing countries with varying degrees of openness and degrees of institutional development. Chapter 1: Introduction This chapter will provide an overview of the BPM6 CG. It will explain why a compilation guide is needed and will give a summary of the major changes in structure and content between the BPM5 CG and the BPM6 CG. It will also outline the scope and organization of the BPM6 CG. SECTION A: OVERVIEW OF MAIN DATA SOURCES Chapter 2: Surveys of Businesses 5 This chapter will discuss the range of surveys of enterprises that may be developed to collect data for compiling balance of payments and IIP statistics. It will highlight the relative benefits of conducting surveys and the relative costs. The chapter will discuss the process for conducting surveys of enterprises including the following: Questionnaire design Frame design Sample selection Data validation Relevant sections will cover surveys to collect information on financial transactions, surveys of international merchandise trade, surveys of services transactions, and surveys of direct and portfolio investment. Chapter 3: Household Budget Surveys and other collections from households 6 This chapter will discuss the data collection from households. In particular, it will review data collection from household budget surveys, surveys of travelers (including visitor and tourism expenditure surveys), and surveys of individuals who cross the border of their country of residence to work or for other purposes. 5 Would include surveys of local enterprises groups and enterprises. 6 Households include individuals as members of households.

17 17 Chapter 4: International Transactions Reporting System This chapter will discuss the uses of the data from the international transactions reporting system (ITRS) in compiling balance of payments statistics. It will provide an overview of the establishment of an ITRS and will review the advantages of an ITRS compared with other data sources. It will also examine the relative disadvantages of an ITRS and the potential departures from balance of payments concepts and principles. Thus, the chapter will examine the key valuation issues relating to ITRS data and how these issues may be resolved. In particular, issues of coverage, timing and valuation will be discussed. Chapter 5: International Merchandise Trade Statistics This chapter will examine data collection from international merchandise trade statistics. It will review the differences between the concepts used in International Merchandise Trade Statistics: Concepts and Definitions and BPM6 concepts, and will discuss the techniques used to adjust the IMTS data to BPM6 concepts. In addition to estimating the goods component of the balance of payments statistics, the chapter will examine how merchandise trade statistics may be used to estimate other components, such as foreign direct investment. Chapter 6: Data from official and administrative sources This chapter will examine the various administrative and official sources that could be used as sources of data such as government accounts, external debt statistics, and central bank statistics. It will also examine the data collected by international organizations such as the Bank for International Settlement s international banking statistics, the World Bank s external debt statistics, and the IMF s international financial statistics. It will outline the various differences in concepts and how these differences may be overcome. Chapter 7: The IMF s global data collection initiatives The chapter will discuss the two global survey initiatives being coordinated by the IMF: the Coordinated Direct Investment Survey (CDIS) and the Coordinated Portfolio Investment Survey (CPIS) and briefly explain how the data could be used in balance of payments and IIP compilation. In regard to the CDIS, it will discuss compilation issues such as the choice of statistical unit (local enterprise groups or enterprises), the identification of ultimate controlling parent, the use of book value in estimating the market value of unlisted equity, and methods of estimating earnings and reinvested earnings on a current basis.

18 18 SECTION B: COMPILATION PRACTICES Chapter 8: Compiling the Balance of Payments and International Investment Position Statistics An Overview This chapter will provide an overview of the compilation of the balance of payments and IIP statement. It will examine broad issues relating to estimation and the various techniques that could be used when data are unavailable or deficient. This chapter will also examine the design and management of the BOP statistical process. The statistical process consists of extraction of data from data sources, estimation of certain data, preparation of a BOP worksheet, verification of data in the worksheet. It will also examine the techniques used to check the consistency of the items within the dataset and across datasets (previously examined in Chapter 20 of BPM5 CG). It will also examine a range of issues that a compiler could face such as multiple exchange rates, residency issues, the use of foreign currency for domestic transactions (dollarization, euroization), and regional arrangements. It also will address direct investment compilation issues. Chapter 9: The International Investment Position a) Positions b) Exchange rate changes c) Other valuation changes This chapter will incorporate material from the publication, International Investment Position: A Guide to Data Sources. It will discuss the estimation of exchange rate changes, estimation of the various valuation changes and the options open to the compiler when these data are not available. It will also discuss deriving an estimate of the opening position when the relevant data are not available. Chapter 10: The Financial Account This chapter will be closely related to the previous chapter. It will be structured according to the functional categories viz. direct investment, portfolio investment, financial derivatives (other than reserves) and employee stock options, other investment, and reserve assets. Much of this chapter will focus on the data sources that could be used to derive the various estimates and the consistency checks that could be employed. It will also examine the use of stock data to estimate transaction flows. Direct investment: In addition to the use of conventional data sources, this chapter will also examine the indirect estimation techniques that could be used and their limitations; for

19 19 example the use of building permits and estimates of construction activity, imports statistics, and tax data. Other investment: Estimating currency and deposits is generally a weak area of the financial accounts. This section will review some country examples and will outline some of the internal consistency checks that could be employed. Reserve assets: the data to compile these estimates are usually derived from official sources. The section will address consistency between the balance of payments and IIP statistics and the central bank accounts, as well as with the IMF financial data. It will include some country examples of the determination of reserve assets (monetary vs. nonmonetary gold, foreign currency holdings, etc.) Chapter 11: Goods This chapter will describe how data on goods transactions could be derived from the primary data sources and will also outline how these data should be adjusted to derive estimates of goods based on BPM6 concepts. It will also examine how enterprise survey data could be used to augment the data from the IMTS and the ITRS. Imports of goods are generally recorded on a CIF basis. The chapter will outline the techniques/source data that could be used to adjust the data to the BOP requirement of FOB basis. Merchanting is now recorded under goods. This chapter will examine some of the data sources that could be used to derive estimates of goods under merchanting. Chapter 12: Services The chapter will examine separately the items that are derived through explicit/direct estimation from source data and the items that are derived implicitly/indirectly. a) Manufacturing services on physical inputs owned by others This section will discuss the data sources that could be used to derive these estimates and will examine some country practices. b) Freight and insurance on imports These two items are closely linked and will therefore be examined together. It will therefore follow from Box 10.3 and Appendix 6c. c) Travel Information on the estimation of travel credits is closely linked to the estimation of visitor/tourism expenditure as recommended by the 2008 International Recommendations for Tourism Statistics. The section will review the differences between travel credits and visitor

20 20 expenditure estimates and will show how the balance of payments item could be derived from the visitor expenditure data. The section will also examine other techniques used to estimate travel such as ITRS data, partner country data, and records of credit card companies. The section will pay particular attention to estimating travel in the absence of border controls and will examine the various methods to estimate the movement of individuals. For travel debits, the section will examine both direct and indirect methods including how household budget survey data could be used to estimate travel debits. d) Insurance services This section will follow from the examples provided in the BPM6. It will expand on these examples by looking at more difficult cases and will review the various sources of data that could be used to estimate insurance services. In particular, the section will explore instances of claims volatility and the implications of improper treatment on the accounts. e) Treatment of intellectual property This section will review treatment of various intellectual property products that may span various balance of payments items as presented in Table 10.4 of the BPM6. It will also clarify the treatment of originals. f) Technical assistance The section will examine the treatment of technical assistance and will follow from Box 10.6 of the BPM6. It will review the various transactions (not all services) that could be derived from a technical assistance program and the various channels through which technical assistance may be delivered. Other sections will examine financial intermediation services indirectly measured and construction. Chapter 13: Primary Income In addition to examining data sources and compilation techniques, this chapter will place relatively greater emphasis on borderline issues, such as differentiating between compensation of employees and services; rent and rental; and taxes and payments for services. Chapter 14: Secondary Income Among other issues, this chapter will examine the measurement of personal transfers. It will borrow from the guidelines on measuring remittances, presented in the Remittances Compilation Guide.

21 21 Chapter 15: Capital Account This chapter will address some borderline issues such as the distinction between capital taxes and current taxes, and capital transfers and current transfers. Chapter 16: A Summary of Insurance Transactions and Positions (possibly an annex) This chapter will review the treatment of the various transactions relating to insurance that have been addressed elsewhere in the BPM6 CG. It will present an example of the (partial) accounts of an insurance enterprise and will show how each transaction and position is reflected in the balance of payments and IIP statistics. Chapter 17: Foreign Direct Investment (possibly an annex) This chapter will provide guidance on the compilation of statistics on foreign direct investment. Areas that will be addressed include presentation of model survey forms for collecting FDI data from enterprises, discussion of the treatment of fellow enterprises, identification of Ultimate Controlling Parents, and statistical unit issues (enterprises versus local enterprise groups). Chapter 18: Exceptional Financing This chapter will discuss the rearrangement of the items in the standard presentation and will review the range of exceptional financing transactions, with examples. Chapter 19: The accessibility and serviceability of the statistics This chapter will address quality issues relating to Dimension Four and Dimension Five of the Data Quality Assessment Framework. It will therefore examine issues such as the periodicity and timeliness of the statistics, revision policy, and the dissemination of data and metadata. This chapter will also review the presentational changes introduced in the BPM6. Appendices Appendix I Appendix II BPM6 Conversion Matrix and Explanatory Notes Linkages within the international accounts Appendix III Linkages with other macroeconomic datasets Appendix IV Model survey forms and other data collection instruments